Introducing gpu::CommandBufferId as a distinct, IdType<...>-based type.

This CL generalizes the pattern used by content::SavePackageId and
cc::SurfaceId and puts it into //base/id_type.h.  Using this pattern for
gpu::CommandBufferId should hopefully help avoid bugs like the mixup
fixed by https://crrev.com/365437.

BUG=514815, 565545
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=wolenetz@chromium.org, sky@chromium.org

Review URL: https://codereview.chromium.org/1548443002

Cr-Commit-Position: refs/heads/master@{#375620}

cc/layers/texture_layer_unittest.cc

@@ -58,7 +58,8 @@ gpu::Mailbox MailboxFromChar(char value) {
 }
 
 gpu::SyncToken SyncTokenFromUInt(uint32_t value) {
-  return gpu::SyncToken(gpu::CommandBufferNamespace::GPU_IO, 0, 0x123, value);
+  return gpu::SyncToken(gpu::CommandBufferNamespace::GPU_IO, 0,
+                        gpu::CommandBufferId::FromUnsafeValue(0x123), value);
 }
 
 class MockLayerTreeHost : public LayerTreeHost {
@@ -145,8 +146,14 @@ struct CommonMailboxObjects {
   explicit CommonMailboxObjects(SharedBitmapManager* manager)
       : mailbox_name1_(MailboxFromChar('1')),
         mailbox_name2_(MailboxFromChar('2')),
-        sync_token1_(gpu::CommandBufferNamespace::GPU_IO, 123, 0x234, 1),
-        sync_token2_(gpu::CommandBufferNamespace::GPU_IO, 123, 0x234, 2) {
+        sync_token1_(gpu::CommandBufferNamespace::GPU_IO,
+                     123,
+                     gpu::CommandBufferId::FromUnsafeValue(0x234),
+                     1),
+        sync_token2_(gpu::CommandBufferNamespace::GPU_IO,
+                     123,
+                     gpu::CommandBufferId::FromUnsafeValue(0x234),
+                     2) {
     release_mailbox1_ = base::Bind(&MockMailboxCallback::Release,
                                    base::Unretained(&mock_callback_),
                                    mailbox_name1_);

cc/output/gl_renderer_unittest.cc

@@ -2185,7 +2185,8 @@ TEST_F(GLRendererTest, OverlaySyncTokensAreProcessed) {
                     viewport_rect, gfx::Transform());
   root_pass->has_transparent_background = false;
 
-  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0, 0x123, 29);
+  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
+                            gpu::CommandBufferId::FromUnsafeValue(0x123), 29);
   TextureMailbox mailbox =
       TextureMailbox(gpu::Mailbox::Generate(), sync_token, GL_TEXTURE_2D,
                      gfx::Size(256, 256), true);

cc/resources/resource_provider_unittest.cc

@@ -121,7 +121,8 @@ class TextureStateTrackingContext : public TestWebGraphicsContext3D {
 
   void genSyncToken(GLuint64 fence_sync, GLbyte* sync_token) override {
     gpu::SyncToken sync_token_data(gpu::CommandBufferNamespace::GPU_IO, 0,
-                                   0x123, fence_sync);
+                                   gpu::CommandBufferId::FromUnsafeValue(0x123),
+                                   fence_sync);
     sync_token_data.SetVerifyFlush();
     memcpy(sync_token, &sync_token_data, sizeof(sync_token_data));
   }
@@ -199,7 +200,8 @@ class ResourceProviderContext : public TestWebGraphicsContext3D {
 
   void genSyncToken(GLuint64 fence_sync, GLbyte* sync_token) override {
     gpu::SyncToken sync_token_data(gpu::CommandBufferNamespace::GPU_IO, 0,
-                                   0x123, fence_sync);
+                                   gpu::CommandBufferId::FromUnsafeValue(0x123),
+                                   fence_sync);
     sync_token_data.SetVerifyFlush();
     // Commit the produceTextureCHROMIUM calls at this point, so that
     // they're associated with the sync point.
@@ -2835,7 +2837,8 @@ class ResourceProviderTestTextureMailboxGLFilters
         use_image_texture_targets_));
 
     unsigned texture_id = 1;
-    gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0, 0x12,
+    gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
+                              gpu::CommandBufferId::FromUnsafeValue(0x12),
                               0x34);
     unsigned target = GL_TEXTURE_2D;
     const GLuint64 current_fence_sync = context->GetNextFenceSync();
@@ -2978,7 +2981,8 @@ TEST_P(ResourceProviderTest, TextureMailbox_GLTextureExternalOES) {
       gpu_memory_buffer_manager_.get(), NULL, 0, 1,
       use_gpu_memory_buffer_resources_, use_image_texture_targets_));
 
-  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0, 0x12, 0x34);
+  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
+                            gpu::CommandBufferId::FromUnsafeValue(0x12), 0x34);
   const GLuint64 current_fence_sync = context->GetNextFenceSync();
   unsigned target = GL_TEXTURE_EXTERNAL_OES;
 
@@ -3047,7 +3051,8 @@ TEST_P(ResourceProviderTest,
       gpu_memory_buffer_manager_.get(), NULL, 0, 1,
       use_gpu_memory_buffer_resources_, use_image_texture_targets_));
 
-  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0, 0x12, 0x34);
+  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
+                            gpu::CommandBufferId::FromUnsafeValue(0x12), 0x34);
   const GLuint64 current_fence_sync = context->GetNextFenceSync();
   unsigned target = GL_TEXTURE_2D;
 

cc/resources/video_resource_updater_unittest.cc

@@ -182,8 +182,9 @@ class VideoResourceUpdaterTest : public testing::Test {
     gpu::Mailbox mailbox;
     mailbox.name[0] = 51;
 
-    const gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
-                                    0x123, 7);
+    const gpu::SyncToken sync_token(
+        gpu::CommandBufferNamespace::GPU_IO, 0,
+        gpu::CommandBufferId::FromUnsafeValue(0x123), 7);
     scoped_refptr<media::VideoFrame> video_frame =
         media::VideoFrame::WrapNativeTexture(
             media::PIXEL_FORMAT_ARGB,
@@ -219,8 +220,9 @@ class VideoResourceUpdaterTest : public testing::Test {
     for (int i = 0; i < kPlanesNum; ++i) {
       mailbox[i].name[0] = 50 + 1;
     }
-    const gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
-                                    0x123, 7);
+    const gpu::SyncToken sync_token(
+        gpu::CommandBufferNamespace::GPU_IO, 0,
+        gpu::CommandBufferId::FromUnsafeValue(0x123), 7);
     const unsigned target = GL_TEXTURE_RECTANGLE_ARB;
     scoped_refptr<media::VideoFrame> video_frame =
         media::VideoFrame::WrapYUV420NativeTextures(

cc/test/render_pass_test_utils.cc

@@ -159,7 +159,8 @@ void AddOneOfEveryQuadType(RenderPass* to_pass,
   const float vertex_opacity[] = {1.0f, 1.0f, 1.0f, 1.0f};
 
   static const gpu::SyncToken kSyncTokenForMailboxTextureQuad(
-      gpu::CommandBufferNamespace::GPU_IO, 0, 0x123, 30);
+      gpu::CommandBufferNamespace::GPU_IO, 0,
+      gpu::CommandBufferId::FromUnsafeValue(0x123), 30);
   *sync_token_for_mailbox_tebxture = kSyncTokenForMailboxTextureQuad;
 
   ResourceId resource1 = resource_provider->CreateResource(

cc/test/test_web_graphics_context_3d.cc

@@ -655,8 +655,8 @@ GLuint64 TestWebGraphicsContext3D::insertFenceSync() {
 
 void TestWebGraphicsContext3D::genSyncToken(GLuint64 fence_sync,
                                             GLbyte* sync_token) {
-  gpu::SyncToken sync_token_data(gpu::CommandBufferNamespace::GPU_IO, 0, 0,
-                                 fence_sync);
+  gpu::SyncToken sync_token_data(gpu::CommandBufferNamespace::GPU_IO, 0,
+                                 gpu::CommandBufferId(), fence_sync);
   sync_token_data.SetVerifyFlush();
   memcpy(sync_token, &sync_token_data, sizeof(sync_token_data));
 }

cc/trees/layer_tree_host_perftest.cc

@@ -297,7 +297,8 @@ class BrowserCompositorInvalidateLayerTreePerfTest
     scoped_ptr<SingleReleaseCallback> callback = SingleReleaseCallback::Create(
         base::Bind(&EmptyReleaseCallback));
 
-    gpu::SyncToken next_sync_token(gpu::CommandBufferNamespace::GPU_IO, 0, 1,
+    gpu::SyncToken next_sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
+                                   gpu::CommandBufferId::FromUnsafeValue(1),
                                    next_fence_sync_);
     next_sync_token.SetVerifyFlush();
     TextureMailbox mailbox(gpu_mailbox, next_sync_token, GL_TEXTURE_2D);

components/mus/gles2/command_buffer_driver.cc

@@ -53,7 +53,7 @@ CommandBufferDriver::Client::~Client() {}
 
 CommandBufferDriver::CommandBufferDriver(
     gpu::CommandBufferNamespace command_buffer_namespace,
-    uint64_t command_buffer_id,
+    gpu::CommandBufferId command_buffer_id,
     gfx::AcceleratedWidget widget,
     scoped_refptr<GpuState> gpu_state)
     : command_buffer_namespace_(command_buffer_namespace),
@@ -450,7 +450,7 @@ void CommandBufferDriver::OnFenceSyncRelease(uint64_t release) {
 
 bool CommandBufferDriver::OnWaitFenceSync(
     gpu::CommandBufferNamespace namespace_id,
-    uint64_t command_buffer_id,
+    gpu::CommandBufferId command_buffer_id,
     uint64_t release) {
   DCHECK(CalledOnValidThread());
   DCHECK(IsScheduled());

components/mus/gles2/command_buffer_driver.h

@@ -16,6 +16,7 @@
 #include "base/time/time.h"
 #include "gpu/command_buffer/common/capabilities.h"
 #include "gpu/command_buffer/common/command_buffer.h"
+#include "gpu/command_buffer/common/command_buffer_id.h"
 #include "gpu/command_buffer/common/constants.h"
 #include "mojo/public/cpp/bindings/array.h"
 #include "mojo/public/cpp/system/buffer.h"
@@ -52,7 +53,7 @@ class CommandBufferDriver : base::NonThreadSafe {
     virtual void UpdateVSyncParameters(int64_t timebase, int64_t interval) = 0;
   };
   CommandBufferDriver(gpu::CommandBufferNamespace command_buffer_namespace,
-                      uint64_t command_buffer_id,
+                      gpu::CommandBufferId command_buffer_id,
                       gfx::AcceleratedWidget widget,
                       scoped_refptr<GpuState> gpu_state);
   ~CommandBufferDriver();
@@ -81,7 +82,7 @@ class CommandBufferDriver : base::NonThreadSafe {
   gpu::CommandBufferNamespace GetNamespaceID() const {
     return command_buffer_namespace_;
   }
-  uint64_t GetCommandBufferID() const { return command_buffer_id_; }
+  gpu::CommandBufferId GetCommandBufferID() const { return command_buffer_id_; }
   gpu::SyncPointOrderData* sync_point_order_data() {
     return sync_point_order_data_.get();
   }
@@ -113,13 +114,13 @@ class CommandBufferDriver : base::NonThreadSafe {
                                const base::TimeDelta interval);
   void OnFenceSyncRelease(uint64_t release);
   bool OnWaitFenceSync(gpu::CommandBufferNamespace namespace_id,
-                       uint64_t command_buffer_id,
+                       gpu::CommandBufferId command_buffer_id,
                        uint64_t release);
   void OnParseError();
   void OnContextLost(uint32_t reason);
 
   const gpu::CommandBufferNamespace command_buffer_namespace_;
-  const uint64_t command_buffer_id_;
+  const gpu::CommandBufferId command_buffer_id_;
   gfx::AcceleratedWidget widget_;
   scoped_ptr<Client> client_;
   scoped_ptr<gpu::CommandBufferService> command_buffer_;

components/mus/gles2/command_buffer_impl.cc

@@ -189,7 +189,8 @@ void CommandBufferImpl::InitializeOnGpuThread(
         void(mojom::CommandBufferInitializeResultPtr)>& callback) {
   DCHECK(!driver_);
   driver_.reset(new CommandBufferDriver(
-      gpu::CommandBufferNamespace::MOJO, ++g_next_command_buffer_id,
+      gpu::CommandBufferNamespace::MOJO,
+      gpu::CommandBufferId::FromUnsafeValue(++g_next_command_buffer_id),
       gfx::kNullAcceleratedWidget, gpu_state_));
   driver_->set_client(make_scoped_ptr(new CommandBufferDriverClientImpl(this)));
   client_ = mojo::MakeProxy(client.PassInterface());
@@ -199,7 +200,8 @@ void CommandBufferImpl::InitializeOnGpuThread(
   if (result) {
     initialize_result = mojom::CommandBufferInitializeResult::New();
     initialize_result->command_buffer_namespace = driver_->GetNamespaceID();
-    initialize_result->command_buffer_id = driver_->GetCommandBufferID();
+    initialize_result->command_buffer_id =
+        driver_->GetCommandBufferID().GetUnsafeValue();
     initialize_result->capabilities = driver_->GetCapabilities();
   }
   gpu_state_->control_task_runner()->PostTask(

components/mus/gles2/command_buffer_local.cc

@@ -323,7 +323,7 @@ gpu::CommandBufferNamespace CommandBufferLocal::GetNamespaceID() const {
   return gpu::CommandBufferNamespace::MOJO_LOCAL;
 }
 
-uint64_t CommandBufferLocal::GetCommandBufferID() const {
+gpu::CommandBufferId CommandBufferLocal::GetCommandBufferID() const {
   DCHECK(CalledOnValidThread());
   return driver_->GetCommandBufferID();
 }
@@ -421,9 +421,10 @@ void CommandBufferLocal::MakeProgressAndUpdateState() {
 
 void CommandBufferLocal::InitializeOnGpuThread(base::WaitableEvent* event,
                                                bool* result) {
-  driver_.reset(new CommandBufferDriver(gpu::CommandBufferNamespace::MOJO_LOCAL,
-                                        ++g_next_command_buffer_id, widget_,
-                                        gpu_state_));
+  driver_.reset(new CommandBufferDriver(
+      gpu::CommandBufferNamespace::MOJO_LOCAL,
+      gpu::CommandBufferId::FromUnsafeValue(++g_next_command_buffer_id),
+      widget_, gpu_state_));
   const size_t kSharedStateSize = sizeof(gpu::CommandBufferSharedState);
   void* memory = nullptr;
   mojo::ScopedSharedBufferHandle duped;

components/mus/gles2/command_buffer_local.h

@@ -19,6 +19,7 @@
 #include "components/mus/public/interfaces/command_buffer.mojom.h"
 #include "gpu/command_buffer/client/gpu_control.h"
 #include "gpu/command_buffer/common/command_buffer.h"
+#include "gpu/command_buffer/common/command_buffer_id.h"
 #include "gpu/command_buffer/common/command_buffer_shared.h"
 #include "ui/gfx/geometry/size.h"
 #include "ui/gfx/gpu_memory_buffer.h"
@@ -88,7 +89,7 @@ class CommandBufferLocal : public gpu::CommandBuffer,
   bool IsGpuChannelLost() override;
   void EnsureWorkVisible() override;
   gpu::CommandBufferNamespace GetNamespaceID() const override;
-  uint64_t GetCommandBufferID() const override;
+  gpu::CommandBufferId GetCommandBufferID() const override;
   int32_t GetExtraCommandBufferData() const override;
   uint64_t GenerateFenceSyncRelease() override;
   bool IsFenceSyncRelease(uint64_t release) override;

content/browser/download/save_types.h

@@ -13,16 +13,16 @@
 #include <vector>
 
 #include "base/files/file_path.h"
-#include "content/common/id_type.h"
+#include "gpu/command_buffer/common/id_type.h"
 #include "url/gurl.h"
 
 namespace content {
 
 class SavePackage;
-using SavePackageId = IdType32<SavePackage>;
+using SavePackageId = gpu::IdType32<SavePackage>;
 
 class SaveItem;
-using SaveItemId = IdType32<SaveItem>;
+using SaveItemId = gpu::IdType32<SaveItem>;
 
 // Map from save_item_id into final file path.
 using FinalNamesMap = std::map<SaveItemId, base::FilePath>;

content/common/cc_messages_unittest.cc

@@ -617,10 +617,12 @@ TEST_F(CCMessagesTest, UnusedSharedQuadStates) {
 TEST_F(CCMessagesTest, Resources) {
   IPC::Message msg(1, 2, IPC::Message::PRIORITY_NORMAL);
   gfx::Size arbitrary_size(757, 1281);
-  gpu::SyncToken arbitrary_token1(gpu::CommandBufferNamespace::GPU_IO, 0, 0x123,
+  gpu::SyncToken arbitrary_token1(gpu::CommandBufferNamespace::GPU_IO, 0,
+                                  gpu::CommandBufferId::FromUnsafeValue(0x123),
                                   71234838);
   arbitrary_token1.SetVerifyFlush();
-  gpu::SyncToken arbitrary_token2(gpu::CommandBufferNamespace::GPU_IO, 0, 0x123,
+  gpu::SyncToken arbitrary_token2(gpu::CommandBufferNamespace::GPU_IO, 0,
+                                  gpu::CommandBufferId::FromUnsafeValue(0x123),
                                   53589793);
   arbitrary_token2.SetVerifyFlush();
 

content/common/gpu/client/command_buffer_proxy_impl.cc

@@ -20,6 +20,7 @@
 #include "content/common/view_messages.h"
 #include "gpu/command_buffer/client/gpu_memory_buffer_manager.h"
 #include "gpu/command_buffer/common/cmd_buffer_common.h"
+#include "gpu/command_buffer/common/command_buffer_id.h"
 #include "gpu/command_buffer/common/command_buffer_shared.h"
 #include "gpu/command_buffer/common/gpu_memory_allocation.h"
 #include "gpu/command_buffer/common/sync_token.h"
@@ -31,8 +32,9 @@ namespace content {
 
 namespace {
 
-uint64_t CommandBufferProxyID(int channel_id, int32_t route_id) {
-  return (static_cast<uint64_t>(channel_id) << 32) | route_id;
+gpu::CommandBufferId CommandBufferProxyID(int channel_id, int32_t route_id) {
+  return gpu::CommandBufferId::FromUnsafeValue(
+      (static_cast<uint64_t>(channel_id) << 32) | route_id);
 }
 
 }  // namespace
@@ -538,7 +540,7 @@ gpu::CommandBufferNamespace CommandBufferProxyImpl::GetNamespaceID() const {
   return gpu::CommandBufferNamespace::GPU_IO;
 }
 
-uint64_t CommandBufferProxyImpl::GetCommandBufferID() const {
+gpu::CommandBufferId CommandBufferProxyImpl::GetCommandBufferID() const {
   return command_buffer_id_;
 }
 
@@ -602,8 +604,9 @@ void CommandBufferProxyImpl::SignalSyncToken(const gpu::SyncToken& sync_token,
 bool CommandBufferProxyImpl::CanWaitUnverifiedSyncToken(
     const gpu::SyncToken* sync_token) {
   // Can only wait on an unverified sync token if it is from the same channel.
-  const uint64_t token_channel = sync_token->command_buffer_id() >> 32;
-  const uint64_t channel = command_buffer_id_ >> 32;
+  const uint64_t token_channel =
+      sync_token->command_buffer_id().GetUnsafeValue() >> 32;
+  const uint64_t channel = command_buffer_id_.GetUnsafeValue() >> 32;
   if (sync_token->namespace_id() != gpu::CommandBufferNamespace::GPU_IO ||
       token_channel != channel) {
     return false;

content/common/gpu/client/command_buffer_proxy_impl.h

@@ -22,6 +22,7 @@
 #include "base/observer_list.h"
 #include "gpu/command_buffer/client/gpu_control.h"
 #include "gpu/command_buffer/common/command_buffer.h"
+#include "gpu/command_buffer/common/command_buffer_id.h"
 #include "gpu/command_buffer/common/command_buffer_shared.h"
 #include "gpu/command_buffer/common/gpu_memory_allocation.h"
 #include "ipc/ipc_listener.h"
@@ -121,7 +122,7 @@ class CommandBufferProxyImpl
   bool IsGpuChannelLost() override;
   void EnsureWorkVisible() override;
   gpu::CommandBufferNamespace GetNamespaceID() const override;
-  uint64_t GetCommandBufferID() const override;
+  gpu::CommandBufferId GetCommandBufferID() const override;
   int32_t GetExtraCommandBufferData() const override;
   uint64_t GenerateFenceSyncRelease() override;
   bool IsFenceSyncRelease(uint64_t release) override;
@@ -229,7 +230,7 @@ class CommandBufferProxyImpl
   // |*this| is owned by |*channel_| and so is always outlived by it, so using a
   // raw pointer is ok.
   GpuChannelHost* channel_;
-  const uint64_t command_buffer_id_;
+  const gpu::CommandBufferId command_buffer_id_;
   const int32_t route_id_;
   const int32_t stream_id_;
   uint32_t flush_count_;

content/common/gpu/client/gpu_context_tests.h

@@ -69,9 +69,8 @@ CONTEXT_TEST_F(SignalTest, InvalidSignalSyncTokenTest) {
 
   // Signalling something that doesn't exist should run the callback
   // immediately.
-  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO,
-                            0,
-                            1297824234,
+  gpu::SyncToken sync_token(gpu::CommandBufferNamespace::GPU_IO, 0,
+                            gpu::CommandBufferId::FromUnsafeValue(1297824234),
                             9123743439);
   TestSignalSyncToken(sync_token);
 };